The standard treatment for malignant glioblastoma multiforme (GBM) includes maximal surgical tumor resection followed by radiation therapy and adjuvant chemotherapy. However, recurrence rates of GBM and the associated patient mortality are nearly 100%. Despite the key role of tumor resection in clinical GBM therapy, most of the pre-clinical studies focus on treating solid intact intracranial GBM tumors without mimicking the clinical scenario of surgical resection. Therefore, implementation of tumor resection in mouse models that recapitulate the clinical disease features are critical in developing clinically translatable therapies for GBM. In the proposed studies, we will first create and characterize different GBM resection models using patient derived CD133+ GBMs based on 3 different phenotypes (invasive, semi-invasive and nodular). While resection of primary tumor has shown clinical benefit, systemically delivered chemotherapeutic agents or direct injection of viruses and placement of carmustine (BCNU) wafers in tumor resection cavities has provided very limited additional benefit. Based on our recent findings that encapsulation of mesenchymal stem cells (MSC) is necessary to prevent rapid wash- out of stem cells post-transplantation in the tumor resection cavity, we will encapsulate human MSC engineered to express in vitro and in vivo imaging markers into synthetic extracellular matrices (sECMs) and evaluate them for their retention, survival and tumor homing in mouse resection models of GBM with different phenotypes. Based on our preliminary studies which indicate that a novel anti-tumor agent consisting of a secretable version of epidermal growth factor receptor targeted nanobody fused to tumor necrosis factor apoptosis inducing ligand (Enb-TRAIL), induces GBM cell killing in both TRAIL resistant and sensitive GBMs, encapsulated MSC-Enb- TRAIL will be tested in different GBM models of resection. To ensure the safety of our approach, we will ultimately engineer human MSC-Enb-TRAIL to express HSV-thymidine kinase (TK), an activatable cellular suicide gene that will allow us to selectively eradicate MSC post-GBM treatment. The incorporation of genetically engineered fluorescent and bioluminescent imaging (BLI) markers into MSC and GBMs will allow us to follow GBM cell invasion, fate of MSC and pharmacokinetics of therapeutic proteins and their efficacy by in vivo BLI, intravital microscopy (IVM), magnetic resonance imaging (MRI) and positron emission tomography (PET) and thus to fine tune the proposed approaches. Once validated, we will initiate a clinical study in which at the time of brain tumor surgery, the main tumor mass will be removed and sECMs encapsulated MSC will be introduced to target a broad spectrum of remaining tumor cells and micro-invasive tumor deposits in the brain. This will have a major impact in saving the lives of many brain cancer patients.